The tool that
refused to hurry.

PrologueA crowded room

The world did not need another issue tracker.

It already had dozens. And every one of them was slightly wrong in the same tired way: built for the person who reads the report, not the person who does the work. Slow to load. Loud with options. A door that opened in the direction you didn't expect.

Karri Saarinen was a principal designer at Airbnb, spending his days inside one of those tools and thinking the same quiet thought on repeat. This could be better. Not revolutionary. Not disruptive. Just better, the way a good chair is better than a bad one.

That thought is where this story starts. Everything after it is a series of small refusals.

2018Three friends, one complaint

They had each lived the same problem from a different chair.

Karri Saarinen, Jori Lallo and Tuomas Artman met in the Helsinki startup scene long before any of this. Karri and Jori had already built a company together: Kippt, a bookmarking tool that went through Y Combinator in 2012 and was acquired by Coinbase two years later. Tuomas had gone to San Francisco to work at Groupon, then Uber, where he architected mobile systems used by thousands of engineers.

So when Jori took a sabbatical in 2018 and pitched the other two over a beer, they weren't guessing. A designer, an engineer, and a founder had each watched the same thing happen in three different companies: teams had evolved past rigid Agile rituals, but the tools hadn't moved. If the tracker didn't work for designers, or engineers, or even managers, who exactly was it working for?

They decided to build the one they wished they'd had. Fast. Opinionated. Beautiful. Built for people like themselves.

2019The single customer

Everyone says strip it down to one feature. They stripped it down to one customer.

The accepted wisdom for a new product is to ship the smallest possible thing and let the market tell you what to build. Linear's founders looked at a market already full of half-finished tools and drew the opposite conclusion. In a crowded room, a minimum product doesn't get noticed. It has to be better on the first day it's used, or it never gets a second.

Karri later wrote this up as an argument against the startup MVP as usually practised: in a competitive category, build a complete, genuinely good product before you launch, not after.

What they narrowed instead was who it was for. Not enterprises. Not everyone. Small, fast product teams who cared about craft, people who looked a lot like the three of them. They bootstrapped at first, and the earliest users were friends at other startups, whose problem they understood because it had been their own problem, every day, for years.

2019Something invisible

Before anyone could see the product, one of them built a foundation.

This is my favourite part of the story. Before launch, Tuomas spent months on the sync engine. Work no user would ever look at, except every single time they pressed a key.

Most web apps are a form that sends data to a server and waits. Linear keeps a real database inside your browser. Every action happens locally, instantly, and syncs in the background over a socket. No spinners. No waiting for a round trip to find out whether the thing you just did actually happened. To a business, this looked like waste: half a year on plumbing while competitors shipped features. To the people who would later use it, it was the entire product, the reason the app felt like it was made of something different.

It was a foundation poured before anyone had decided what the house would look like. Which, if you think about it, is the only moment a foundation can be poured honestly.

A chart can tell you what happened. It cannot tell you what is true. These are not the same thing.

They didn't ask the numbers what to do next. There weren't enough numbers to ask. They leaned instead on a quiet certainty about what felt right. Harder to defend, and truer to be right about.

2019 to 2021Letting the world in slowly

Ten thousand people asked for an invite. They didn't open the gates.

Linear launched in April 2019 with a Medium post. Around ten thousand people joined the waitlist. And the founders, sitting on exactly the kind of demand most startups pray for, chose not to convert it. The product stayed invite-only, and they let people in slowly enough to remember their names, enough to hear the same complaint three times before deciding it was real.

They said no to money, and no to more people, for longer than seemed wise. When a Sequoia partner reached out because people she trusted kept mentioning Linear, they took a seed round, and stayed small anyway. Roughly two years in, they had their first profitable month with seventeen people and more than a thousand startups paying. Not because they couldn't hire. Because a company shaped too early by too many hands becomes something nobody actually meant to build.

PrincipleThe line they never crossed

Nothing that made a manager's report tidier at the cost of the person doing the work.

Every tool in this category eventually faces the same pressure. The person who approves the invoice is not the person who lives inside the product eight hours a day, and the two want different things. Most tools quietly side with whoever pays.

Linear drew a line early and kept it. The one who lived in the tool was the only person they were building for. Whoever approved the expense was somebody else's concern. Sales, maybe, but not product. They wrote down how they believed teams should work, called it the Linear Method, and declined to be everything to everyone. Opinions cost you some customers. They also make the product coherent.

PracticeWater through stones

None of this resembled speed.

It moved the way water moves through stones: no hurry, no stopping. Small teams gathered around a problem and dissolved once it was solved. Every week, a short and honest changelog of what actually shipped, instead of a long roadmap promising everything and meaning nothing.

A single unfixed bug outweighed any new feature, because a promise to someone who is already a customer matters more than a promise to someone who has never set foot inside. Quality wasn't a phase to be reached after growth. It was the growth strategy.

Each choice was a small, quiet refusal of the easy and visible reward, in favour of something slower, harder to prove, and more real.

FairnessA fair accounting

Temperament was not the only ingredient.

It would be dishonest to leave it there. Three founders with Airbnb, Uber and Coinbase on their CVs, a previous Y Combinator exit, and an audience on Twitter large enough to fill a ten-thousand-person waitlist and bring Sequoia to the door. That is not the starting position most people get. Restraint is easier when you can afford it. Saying no to funding is easier when funding is asking you.

But be fair in the other direction too. Pedigree gets you the first thousand signups. It does not make an app feel instant, and it does not keep you saying no for five years while the visible rewards pile up on the other side of the line. The advantage opened the door. What walked through it was the work.

EpilogueTemperament, not method

By the fifth year, it is the only thing still standing.

Today Linear is valued at over a billion dollars, and the teams building OpenAI, Ramp and Vercel run on it. That's the headline. It is also the least interesting part.

If there is a lesson in the story, it isn't a method you can copy. It's closer to a temperament: a willingness to choose the slower, truer thing when nobody is watching and nobody will applaud. No one claps for that in the first year. By the fifth, it's the only thing still standing. Patient and unremarkable, like an old tool your grandparents made by hand and that still fits your grip.